Previous neuroimaging research suggests that age-related decline in the functioning of specific neural systems contributes to decline in cognitive performance, even in the absence of significant disease. Previous imaging studies, however, have typically focused on the magnitude of mean activation, rather than functional connectivity (FC) of activated regions, and have relied on extreme-group designs comparing younger and older adults. Building on the findings from the previous project period, we propose to investigate age-related differences in visual attention with a multimodal imaging methodology comprising both resting-state and task- related FC, as well as white matter integrity as assessed by diffusion tensor imaging (DTI). Each of four experiments will include 90 healthy, community-dwelling adults sampled continuously across the age range of 18-79 years. Critically, the research will use graph theoretical analyses of structural connectivity (SC) and FC, which will provide a unifying methodology and specific metrics (e.g., strength of FC within and between network modules) that can be applied to both structural and functional imaging. Each of the four experiments will involve a visual search task, and behavioral assessments of visual search performance will implement a mathematical model of reaction time that yields separate parameter estimates for the rate of information accumulation towards a decision (drift rate), as distinct from visual encoding and response execution processes (nondecision time). We propose to elicit differential engagement of functional modules, by varying the attentional demands of the search task related to: the enhancement of target information versus suppression of irrelevant but salient information (Exp. 1), the spatial selectivity of attention (Exp. 2), the activation of a motor response (Exp. 3), and the response to visually degraded targets (Exp. 4). Graph theoretical analyses of the structural and functional imaging data will test the overall scientific premise of this project: that aging is associated with a shift in the dynamic balance of within-module and between-module task-related FC (Aim 1), that this pattern is related to the degradation of visual information as reflected in SC and FC of visual sensory regions (Aim 2), and that task-related FC will reflect the participation of specific modules as a function of attentional task demands (Aim 3). We expect an age-related decline in the graph theoretical measures of within-module connectivity, but an increase with age in measures of between- module connectivity. Further, we hypothesize that the increase in between-module FC with age will be more prominent during task performance, relative to a resting-state, reflecting older adults? increased reliance on top- down attention to support task performance. The four studies and associated network analyses, with healthy adults, are designed to improve current theories of neurocognitive aging and provide information relevant for diagnosis and treatment of neurodegenerative disease.